BTW, I think this means your original IPoIB patch that did:

 > +    path->pathrec.mtu            = priv->broadcast->mcmember.mtu;
 > +    path->pathrec.mtu_selector   = IB_SA_GTE;

now needs to do something like

+       path->pathrec.mtu            = max(IB_MTU_256, 
priv->broadcast->mcmember.mtu - 1);
+       path->pathrec.mtu_selector   = IB_SA_GT;

right?

The strict inequality semantics defined by the spec are somewhat more
awkward to actually use :(

 - R.

_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to